CN111326025A - System and method for predicting flight delay based on real-time data - Google Patents

Abstract

Systems and methods for predicting flight delays based on real-time data. The method may include receiving flight plan data representative of a set of current flight plans and receiving monitoring data representative of a set of current aircraft states. The method may further include generating consolidated data representative of a set of remaining flight portions to be flown. The method may also include receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof, and receiving weather data. The method may include generating predicted flight traffic data by performing a simulation of the flight over the duration of time, the simulation based at least in part on the consolidated data, the operational context data, and the weather data. The method may further include generating a user output based at least in part on the predicted flight traffic data.

Description

System and method for predicting flight delay based on real-time data

Technical Field

The present disclosure relates generally to predicting flight traffic (flight flow), and in particular to predicting flight delay using real-time data.

Background

With the increase in air traffic (air traffic) worldwide, there is a need for more efficient use of air navigation systems in air traffic management and control environments, and thereby reduce delays and congestion. Accurate prediction of recent air traffic delays may help identify areas and participants, which may help balance the capacity provided by air traffic control agencies, air navigation service providers, and airports with the capacity required for upcoming air traffic. A typical airline will be interested in knowing the near-term conditions of air traffic and the likely interactions of traffic with airports, airspaces, and other airlines. For example, when the traffic comes from outside the European Civil Aviation Conference (ECAC) area, the european Air Navigation Service Provider (ANSP) may not know the upcoming traffic until approximately 90 minutes before the aircraft enters its airspace.

Conventional systems for predicting delays in real-time may rely on historical data as well as some real-time information extracted from the ANSP service provider, such as actual departure times or flight plans. This real-time information is used in very simple kinematic algorithms that propagate this information in flight plans to obtain estimated times of arrival and thus the predicted delays for these flights. Mathematical models have been proposed that attempt to link certain precursors of delays to certain elements that may be observed from flight (e.g., city pairs, weather, etc.). These models use different machine learning techniques to train the predictive models.

Typical solutions may lack the accuracy and authenticity of the delay data. Many interactions that affect air traffic timing and delay (e.g., sectors, terminal controlled areas (terminals), waits (recesses), runway configurations, etc.) are not considered in the typical model because they are traditionally outside the scope of air traffic management systems. In some cases, a typical solution that only determines whether the delay is above a certain threshold (typically 60 minutes) is reliable. Traditionally, air traffic fast time simulators have been used as a method to analyze historical delay data. However, current commercial air traffic simulators are designed to be used offline during strategic or pre-tactical planning phases as well as during post-operations. These typical simulators cannot be equipped to work with real-time information feeders. Other disadvantages may exist.

Disclosure of Invention

Systems and methods for predicting airport delays based on flight data and ground data are disclosed herein. The system and method may enable a fast time simulator to be used as a near-term situation prediction engine. The system may combine the data sources for all flights in a given area with the flight plan to algorithmically calculate predicted flight traffic. The data may be used to determine a delay, and the predicted delay may be used to provide a recommendation to mitigate the delay. The system can collect and combine real-time information from different feeders to provide accurate predictions of traffic conditions.

In one embodiment, the method includes receiving flight plan data representative of a set of current flight plans. The method also includes receiving monitoring data representative of a set of current aircraft states. The method further comprises generating consolidated data representative of a set of remaining flight portions to be flown, wherein the consolidated data is generated by: pairing each aircraft state (aircraft availability status) with each flight plan (aircraft flight); correcting the respective flight plans based on the respective aircraft states to generate corrected flight plans; and determining the remaining portions of the corrected flight plan to be flown. The method includes receiving operational context data representing an airspace configuration, an airport configuration, or a combination thereof. The method also includes receiving weather data. The method also includes generating predicted flight traffic data by performing a simulation of the flight over the duration of time, wherein the simulation is based at least in part on the consolidated data, the operational context data, and the weather data. The method also includes generating a user output based at least in part on the predicted flight traffic data.

In some embodiments, the method comprises: receiving user input representing a duration and an area of interest (an area of interest), and filtering the set of current flight plans based on the duration and on the area of interest. In some implementations, the area of interest is associated with an airport terminal, with an airport, with multiple airports, or with a geographic area. In some embodiments, each flight plan of the set of flight plans includes: a representation of a route to be followed, an aircraft type, an origin-destination pair, an estimated departure time, an estimated arrival time, or any combination thereof. In some embodiments, the monitoring data includes an automatic dependent broadcast-broadcast (ADS-B) report. In some embodiments, the method comprises: storing at least one ADS-B report; and calculating flight trajectories based on the ADS-B reports, wherein at least one of the corrected flight plans is based on flight trajectories. In some embodiments, the method includes formatting the weather data to be compatible with the simulated architecture. In some embodiments, the weather data is formatted into a weather information file, the operational context data is formatted into waypoint and navigation assistance files, airport files, or both, and the consolidated data is formatted into a flight schedule file. In some embodiments, the simulation is performed using a fast time simulation tool. In some implementations, the user output includes a predicted average delay for the region of interest at a time within the duration, a suggested action to reduce the predicted average delay, or both.

In one embodiment, a system includes at least one processor and at least one memory storing instructions that, when executed by the processor, cause the at least one processor to receive flight plan data representative of a set of current flight plans. The instructions also cause the at least one processor to receive monitoring data representative of a set of current aircraft states. The instructions further cause the at least one processor to generate consolidated data representative of a set of portions of the flight remaining to be flown, the consolidated data generated by: pairing each aircraft state with each flight plan; correcting the respective flight plans based on the respective aircraft states to generate corrected flight plans; and determining the remaining portions of the corrected flight plan to be flown. The instructions also cause the at least one processor to send the merged data to the simulator module.

In some embodiments, the instructions cause the processor to: receive operational context data representative of an airspace configuration, an airport configuration, or a combination thereof, receive weather data, and send the operational context data and the weather data to the simulator module. In some embodiments, the simulator module is configured to generate the predicted flight traffic data by performing a simulation of the flight for the duration of time, wherein the simulation is based at least in part on the merged data, the operational context data, and the weather data, and configured to output at least a portion of the predicted flight traffic data to the output device. In some embodiments, the instructions further cause the processor to receive user input representing a duration and a region of interest, and filter the set of current flight plans based on the duration and based on the region of interest. In some embodiments, the instructions further cause the processor to store at least one ADS-B report in the memory and calculate flight trajectories based on the ADS-B reports, wherein at least one of the corrected flight plans is based on flight trajectories. In some embodiments, the instructions further cause the processor to format the weather data to be compatible with an architecture of the simulator module.

In one embodiment, the method includes receiving flight plan data representative of a set of current flight plans. The method also includes receiving monitoring data representative of a set of current aircraft states. The method also includes generating consolidated data representing a set of remaining flight portions to be flown based on the flight plan data and based on the monitoring data. The method includes receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof. The method also includes receiving weather data. The method also includes iteratively changing parameters of the operational context data by performing a plurality of simulations of flights over a duration of time, and calculating a set of predicted delays for the flights for each iteration. The plurality of simulations are based at least in part on the consolidated data, the operational context data, and the weather data. The method includes identifying a recommendation associated with a parameter to reduce a prediction delay based on a plurality of simulations.

In some embodiments, the method comprises: the method further includes receiving user input indicative of a duration and a region of interest, and filtering the set of current flight plans based on the duration and based on the region of interest. In some implementations, the area of interest is associated with an airport terminal, with an airport, with multiple airports, or with a geographic area. In some embodiments, the method includes generating a user output including a predicted average delay for the region of interest at a time within the duration.

Drawings

FIG. 1 is a block diagram depicting an embodiment of an overall system for predicting flight traffic.

FIG. 2 is a block diagram depicting an embodiment of a system for predicting flight traffic and delay.

FIG. 3 is a block diagram depicting an embodiment of a system for merging flight plan data and surveillance data.

FIG. 4 is a flow chart depicting an embodiment of a method for predicting flight traffic.

FIG. 5 is a graph depicting a comparison of predicted flight traffic and actual flight traffic.

FIG. 6 is a graph depicting the accuracy of the delay prediction as a function of time until the aircraft lands.

FIG. 7 is a flow chart depicting an embodiment of a method for predicting flight traffic.

FIG. 8 is a flow diagram depicting an embodiment of a method for identifying recommendations based on predicted flight traffic.

While the disclosure is susceptible to various modifications and alternative forms, specific embodiments have been shown by way of example in the drawings and will be described in detail herein. It should be understood, however, that the disclosure is not intended to be limited to the particular forms disclosed. On the contrary, the intention is to cover all modifications, equivalents, and alternatives falling within the scope of the disclosure.

Detailed Description

Referring to FIG. 1, an embodiment of an overall system 100 for predicting flight traffic is depicted. The system 100 may include a real-time data provider 110, a simulation system 120, and associated support tools 170.

The real-time data provider 110 may provide real-time data to the simulation system 120. For example, the first provider 112 may provide flight plan data 124, flight data 125, and operational context data 126. The second provider 113 may provide the monitoring data. The third data provider 114 may provide weather data. It should be noted that this particular configuration is for example purposes only. Other combinations including other providers are possible. Examples of data providers may include ADS-B providers (e.g., flight Radar 24 or flight Aware), central traffic flight plan collectors (e.g., Eurocontrol's network managers or airline operations centers), meteorological services (e.g., national oceanic and atmospheric administration operations model archives and distribution systems (NOMADS)), and so forth.

Providers 112 through 114 may be associated with respective queue structures 116 through 118. For example, a first provider 112 may be associated with a first queue structure 116, a second provider 113 may be associated with a second queue structure 117, and a third data provider 114 may be associated with a third queue structure 118. The queue structures 116 to 118 may include: apparatus and protocol for formatting, organizing, queuing, and streaming real-time data. For example, the queue structures 116 through 118 may be implemented using Advanced Message Queuing Protocol (AMQP), Apache Kafka, other Message transport and streaming platforms, or a combination thereof.

The simulation at simulation system 120 may be performed in stages. During the data collection phase 122, flight plan data 124, flight data 125, operational background data 126, surveillance data 127, and weather data 128 may be received and organized.

The flight plan data 124 may represent a set of current flight plans. As used herein, the term "current" means that the flight plan is associated with a flight that has not yet been completed, as opposed to "historical" data that is associated with a completed flight. The flight plan data 124 may be used as a baseline element to provide general information about all current flights in the air traffic system 164 of interest during the time period of interest 162. The purpose of obtaining flight plan data 124 is to provide the information needed to properly simulate all relevant flights during a specified time period of interest 162. The flight plan data 124 may include representations of routes to be followed, aircraft types, origin-destination pairs, estimated departure times, estimated arrival times, or any combination thereof, for each flight plan.

The flight data 125 may include in-flight updates. It may be similar to the monitoring data 127, but may be reported at different intervals. The flight data may provide 4D position updates of the aircraft. In some cases, flight data 125 may be used when monitoring data 127 is not available.

The operational context data 126 may include relevant information related to the airspace and airport environment within the air traffic system 164 of interest. It may take the form of an aeronautical information exchange model (AIXM). As such, operational context data 126 may include information related to: airport area data, airspace structures, organizations and units, points and navigation aids, procedures for constructing waypoints, route and flight restriction information, runway configurations, airport information, and route information, which may be useful for performing accurate simulations of airspace systems.

Surveillance data 127 may include observed flight traffic data, such as radar information or automatic dependent surveillance broadcast (ADS-B) reports. In this way, the monitoring data 127 may represent a set of current aircraft states. As with the flight plan data 124, the term "current" means that the status is associated with a flight that has not yet been completed. The purpose of obtaining the monitoring data 127 is to determine the starting state of each aircraft to be simulated. In addition, the monitoring data 127 may be recorded and stored during the time period of interest 162 and may be used for calibration trajectory calculations.

Weather data 128 may provide weather (e.g., wind) information useful for calculating aircraft trajectories. For example, whether the speed and development of the flight may be affected.

After the data collection phase 122, a data processing phase 130 may be performed. A user input 160 may be received and the user input 160 may include a time period of interest 162 and an air traffic system of interest 164. The focus period 162 may include a start-up time, which in real-time applications will typically be within a threshold of the time to perform the simulation. The threshold may depend on factors such as the reporting frequency of the real-time data provider 110 and other factors (e.g., technical limitations) that may cause delays in the system 100. The focus period 162 may also include a time range, or duration, i.e., the time to be simulated. The air traffic system of interest 164 may include areas of interest that may correspond to different spatial dimensions, such as airports or airport networks in countries, continents, and the like.

The data processing stage 130 may generate customized simulation parameters 132, a simulation background 133, and a simulation input file 134. To generate the customized simulation parameters 132, the data 124-128 may be filtered according to a time period of interest 162 and according to an air traffic system of interest 164. This may limit the simulation to those flights that are active within the air traffic system 164 of interest.

Each flight from flight plan data 124 can be correlated with monitoring data 127. The flight plan data 124 may be corrected using a trajectory intercept routine based on the monitoring data 127. The process may include: determining points in each flight plan for which particular data points (e.g., ADS-B reports) from the monitoring data 127 are appropriate; and to maintain a portion of the flight plan that the aircraft still needs to fly. The consolidated data may replicate the current conditions of the airspace of the air traffic system 164 of interest at the start time of the time period of interest 162.

The merged data may be further combined with operational context data 126 and weather data 128 to generate a simulated context 133. Based on the combination of all the data 124 to 128 and the user input 160, a simulation input file 134 may be generated. The simulation system 120 may use the fast time simulator 140 to perform simulations of flights in the time period of interest 162 within the air traffic system of interest 164. It should be noted that the system 100 allows for a modular design, which means that the machine in which the implementation of the fast time simulator 140 takes place can be independent of the machine that collects and processes the data 124 to 128. This may enable the system 100 to be implemented as a cloud service, where users may avoid direct interaction with the fast time simulator 140, simplifying the use of the system 100.

The results of the simulation performed by the fast time simulator 140 may be analyzed in a post-processing stage 150 to provide information to the user. Outputs that may be generated include: stream output 152, correlation operation metrics output 153, course or tail performance tracking output 154, and system state output 155 in different time ranges. Other outputs may also be generated. In some implementations, the one or more outputs 152-155 may include a predicted average delay for the air traffic system of interest 164 at a time within the time period of interest 162. In some implementations, one or more of the outputs 152-155 may include suggested actions for reducing the predicted average delay. Outputs 152 to 155 may likewise include both the predicted average delay and the proposed action. Other outputs are also possible.

The support tools 170 may include tools to help interested parties view and understand the data generated by the fast time simulator 140. For example, the support tool 170 may include a virtual radar server 172 to provide a visualization of the stream output 152. Other support tools 170 are also possible.

The system 100 may enable the delay to be predicted in real-time at any participant (e.g., airport and airline) involved in the air traffic management. This will permit the airline to plan emergency actions to reduce its latency and improve its punctuality. The system 100 may provide a real-time representation of the location of delay accumulation, delay trends, and delay root causes. Airports may benefit from this system by understanding, based on outputs 152 to 155, which elements of their airport (e.g., particular routes, ground movements, and use of runways) are causing delays. The system 100 may provide delay metrics and indicators (indicators) to a user (e.g., an airline or airport) related to the user's operation (e.g., airport/airspace delays, congestion, probability of waiting (holding), airport/airspace configuration changes, etc.).

The predicted delay provided by the system 100 may provide the airline with a quantification of the impact of some common external factors affecting flight, such as weather (wind, storm, etc.), changes in airspace configuration (e.g., sector closures), and airport congestion. With this information, the airline will be able to know the conditions of the air traffic system before its flight departs or while it is in flight, and which conditions may affect its trajectory.

The airline or airport may also use the system 100 to try different possible scenarios (e.g., request a particular departure/arrival runway, re-plan a selected airline or even aircraft type) based on real-time data obtained from the feeder, and make a new choice to its pilot or ANSP to recover from the delay. Other advantages may exist.

Referring to FIG. 2, an embodiment of a system 200 for predicting flight traffic and delays is depicted. The system 200 may include a first data provider system 202, a second data provider system 204, a third data provider system 206, a fourth data provider system 208, and a fifth data provider system 210. The first data provider system 202 may include flight plan data 124, the second data provider system 204 may include flight data 125, the third data provider system 206 may include operational context data 126, the fourth data provider system 208 may include surveillance data 127, and the fifth data provider system 210 may include weather data 128. In some implementations, some of the data provider systems 202 through 210 can be combined. For example, the flight plan data 124, flight data 125, and operational context data 126 may be associated with a single data provider system as shown in FIG. 1. Other combinations are also possible.

The system 200 may further include a data processing module 220, a simulation module 230, and a post-processing module 240. Although fig. 2 depicts the modules 220, 230, 240 as being distinct, in some implementations, one or more of the modules 220, 230, 240 may be combined. For example, in some embodiments, the functions described with respect to each of modules 220, 230, 240 may be performed by a single computing module. As another example, the data processing module 220 and the post-processing module 240 may be combined, while the simulation module 230 may be remotely accessed. Other combinations are also possible.

Each of the modules 220, 230, 240 may include a respective processor and memory. For example, the data processing module 220 may include a first processor 222 and a first memory 224. The simulation module 230 may include a second processor 232 and a second memory 236. The post-processing module 240 may include a third processor 242 and a third memory 246. As described above, in some implementations, the modules 220, 230, 240 may be combined. For example, various ones of the functions described with respect to the modules 220, 230, 240 may be performed by a single processor and memory. Other combinations are also possible.

The data processing module 220 may be configured to communicate with the data provider systems 202 to 210 via the network 212. Network 212 may include a Local Area Network (LAN), a Wide Area Network (WAN), another type of private or public network, or a combination thereof. In some implementations, the network 212 may include the internet.

The post-processing module 240 may further include an output device 248 for presenting the simulation results to a user. The output device 248 may include a visual display device (e.g., a screen or monitor), an audio device (e.g., a speaker), another type of output device, or a combination thereof.

During operation, the data processing module 220 may receive from the data provider systems 202 to 210 via the network 212: flight plan data 124, flight data 125, operational background data 126, surveillance data 127, and weather data 128. The data processing module 220 may further receive user input 260 representing a duration 262 and a region of interest 264. The data 124-128 may be filtered based on the duration 262 and based on the region of interest 264. As an example, the flight plan data 124 may be limited to flight plans that affect the area of interest 264 during the duration 262. Other flight plans included in the flight plan data 124 may be discarded.

The data processing module 220 may use the flight plan data 124 and the monitoring data 127 to generate consolidated data 268 representing a set of flight portions remaining to be flown, which may be stored in the memory 224. In some embodiments, the flight data 125 may also be used to generate the consolidated data 268, particularly where access to the monitoring data 127 may be limited.

The merged data 268, as well as the operational background data 126 and the weather data 128, may be sent to the simulation module 230. In some embodiments, the merged data 268, operational background data 126, and weather data 128 may be formatted to be compatible with the architecture of the simulation module 230. For example, a particular file format may be used.

The simulation module 230 may receive the merged data 268, the operational background data 126, and the weather data 128, and may generate predicted flight traffic data 270 by performing a simulation of the flight over the duration 262. The simulation may be based at least in part on the merged data 268, the operational background data 126, and the weather data 128. Further, in some implementations, multiple simulations may be performed at the simulation module 230 in order to identify the recommendations 284 to improve one or more parameters associated with air traffic in the area of interest 264 and during the duration 262. For example, the data processing module 220 or the simulation module 230 may iteratively change the parameters of the operational context data 126 by performing multiple simulations of the flight over the duration 262 and, for each iteration, calculate a set of predicted delays 272 for the flight. In one embodiment, the simulation module 230 may be implemented using a fast time simulation tool.

The post-processing module 240 may use the predicted flight traffic data 270 to generate predicted delays 272 associated with aircraft in the area of interest 264, with a group of aircraft, with an airport, or other participants associated with the area of interest 264. User output 280 may be generated. The user output 280 may include an average delay 282 associated with flight in the region of interest 264 over the duration 262 and may also identify a recommendation 284 associated with a parameter selected for iterative change. One of ordinary skill in the art having the benefit of this disclosure will recognize that other outputs are possible.

A benefit of the system 100 is that real-time prediction of delay can be performed with significantly improved accuracy. While existing air traffic prediction systems may rely solely on flight data (wind speed, heading, etc.) and weather data at a single airport to provide predictions, the system 100 enables a more accurate understanding of the source (source) or delay, including operational context data not considered by general air traffic prediction systems. In addition, the system may use a fast time simulator to make predictions, which has typically been used to analyze historical data rather than real-time data. Other advantages may exist.

Referring to FIG. 3, an embodiment of a system 300 for merging flight plan data 124 and surveillance data 127 is depicted. The system 300 may be implemented at the data processing module 220 (shown in fig. 2) to generate the consolidated data 268.

The consolidated data 268 may be generated by pairing individual aircraft states 322, 323, 324 of the set of aircraft states 320 represented by the monitoring data 127 with individual flight plans 312, 313, 314 of the set of current flight plans 310 represented by the flight plan data 124.

In some implementations, each individual aircraft state 322-324 may be stored in memory 224 as a stored state 332, 333, 324. The stored states 332-324 may be used to calculate flight trajectories 342, 343, 344 associated with the flight. In some embodiments, each aircraft state 322-324 corresponds to one or more ADS-B reports, and flight trajectories 342-344 may be calculated using the one or more ADS-B reports.

The respective flight plans 312 to 314 may be corrected based on the respective aircraft states 322 to 324 using the flight trajectories 342 to 344 to generate corrected flight plans 352 to 354. Then, a trajectory intercept method may be employed to determine the remaining portions 362-364 of the corrected flight plans 352-354 to be flown. The merged data 268 may represent a set of remaining flight portions 360 to be flown.

As described above, the system 300 may be implemented by the data processing module 220 to prepare the merged data 268 for simulation. The system 300 is depicted for exemplary purposes only. Those skilled in the art having the benefit of this disclosure will appreciate that additional features and elements may be present to assist in generating the consolidated data 268.

Referring to FIG. 4, an embodiment of a method 400 for predicting flight traffic is depicted. At 402, the method 400 may include determining a user-defined temporal and spatial framework for the simulation. For example, user input 260 including duration 262 and area of interest 264 may be received at data processing module 220. The area of interest 264 may be associated with an airport terminal, with an airport, with multiple airports, or with a geographic area.

At 404, the method 400 may include a data collection phase. During the data collection phase, weather data 128, operational background data 126, surveillance data 127, flight data 125, and flight plan data 124 may be collected. For example, the data processing module 220 may collect the data 124 to 128.

At 406, the method 400 may include formatting the weather data 128 using a Total Airspace and Airport Modeler (TAAM) tool for use with the simulation tool. At 408, the method 400 further includes determining from the operational context data 126 which air traffic data is available. For example, in some cases, the operational context data 126 may be limited or incomplete. By using available operational context data 126 for the determination, the accuracy of the prediction may be improved.

Based on the user-defined temporal and spatial framework determined at 402, logic for the defined air traffic system may be constructed at 410. In other words, the method 400 may determine which airports, aircraft, and parameters should be included in the simulation. Further, simulation parameterization may be performed at 426 to determine what output should be generated for display to the user. For example, the parameters of the simulation may depend on whether the user is paying attention to the average delay of the system, the real-time delay of the aircraft, the suggested actions to mitigate the delay, and so forth.

The method 400 may also include developing (develoop) filter parameters at 412. At 414, the flight plan data 124 may be filtered using these parameters, resulting in only the flight plan corresponding to the temporal and spatial framework defined at 402 remaining for simulation. Then, at 416, each flight plan of the flight plan data 124 can be correlated with the monitoring data 127 and with the flight data 125. Then, at 418, the monitoring data 127 and the flight data 125 may be fit within the 4-dimensional locations of the flight plan data. This may result in consolidated data representing a set of remaining flight portions to be flown. Further, at 420, additional operations (e.g., aircraft type, airport, etc.) may be determined from the flight plan data 124 for simulation. At 422, the entire set of operations to be simulated may be compiled based on the consolidated data generated at 418 and the additional operations determined at 420. At 424, based on the available air traffic data determined at 408 and the operation to be simulated determined at 422, the required air traffic system context data may be determined.

Each of these analog inputs may be formatted into a particular file format 430 that may be used by the analog architecture. For example, weather data 128 may be formatted into a weather information file 432. The operational context data 126, after being compiled into air traffic system context data, may be formatted into waypoint and navigation assistance files 434, airport files 436, or both. The consolidated data representing the operations to be simulated determined at 422 may be formatted into a flight schedule file (flight schedule file) 438. The simulation parameters determined at 426 may be formatted into a simulation parameters file 440. Although these file formats may be used for fast time simulation tools, other formats may be used with other simulation tools. After files 432 through 440 are generated, a simulation may be performed at 450.

Referring to FIG. 5, a graph depicts predicted flight traffic versus actual flight traffic for a single day. The x-axis corresponds to each hour of the day. The left axis corresponds to the number of arrival movements and is represented by the bar graph depicted on the graph. The right axis corresponds to the average delay associated with the arrival motion and is represented by a line graph representing actual data as compared to simulated data. As shown in fig. 5, the simulated data closely follows the real data. A user who obtains simulated data ahead of time will be able to detect, for example at 11 o' clock, that the airport is expected to have a delay of approximately 40 minutes relative to the planned arrival time. In this way, the user will be able to take action accordingly.

Referring to FIG. 6, a graph depicts the accuracy of the delay predictions produced by the systems and methods described herein as a function of time until the aircraft lands. As shown in fig. 6, the system described herein may produce a relatively low delay error even when the prediction is made 135 minutes prior to landing. Based on this accuracy, for example, if the delay at an airport is high according to the prediction, the airline may decide to slow down or accelerate the aircraft to avoid saturation of the arrival at the airport.

Referring to FIG. 7, an embodiment of a method 700 is depicted, which is an embodiment of a method for predicting flight traffic. At 702, method 700 may include receiving flight plan data representative of a set of current flight plans. For example, the flight plan data 124 may be received at the data processing module 220.

At 704, method 700 may further include receiving monitoring data representative of a set of current aircraft states. For example, the monitoring data 127 may be received at the data processing module 220.

At 706, method 700 may also include generating consolidated data representative of a set of remaining flight portions to be flown. The merged data may be generated by: pairing each aircraft state with each flight plan; correcting the respective flight plans based on the respective aircraft states to generate corrected flight plans; and determining the remaining portions of the corrected flight plan to be flown. For example, the consolidated data 268 may be generated at the data processing module 220.

At 708, the method 700 may include receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof. For example, the operational context data 126 may be received at the data processing module 220.

At 710, the method 700 may further include receiving weather data. For example, weather data 128 may be received at data processing module 220.

At 712, the method 700 may further include generating predicted flight traffic data by performing a simulation of the flight over the duration of time, the simulation based at least in part on the merged data, the operational context data, and the weather data. For example, predicted flight traffic data 270 may be generated at the simulation module 230.

At 714, method 700 may include generating a user output based at least in part on the predicted flight traffic data. For example, user output 280 may be generated at post-processing module 240.

Referring to FIG. 8, an embodiment of a method 800 for identifying recommendations based on predicted flight traffic is depicted. At 802, method 800 may include iteratively changing parameters of operational context data by performing a plurality of simulations of flights over a duration of time, and for each iteration, calculating a set of predicted delays for the flights. The plurality of simulations may be based at least in part on consolidated data representing a set of remaining flight portions to be flown, operational background data, and weather data. At 804, method 800 may further include identifying a recommendation associated with the parameter to reduce the prediction delay based on the plurality of simulations.

A benefit of the method 800 is that it enables accurate real-time simulations to be performed to determine suggested actions that reduce latency. Other advantages may exist.

Further, the present disclosure includes embodiments according to the following clauses:

clause 1: a method, the method comprising: receiving flight plan data representative of a set of current flight plans; receiving monitoring data representative of a set of current aircraft states; generating consolidated data representative of a set of remaining flight portions to be flown, said consolidated data being generated by: pairing each aircraft state with each flight plan, correcting the each flight plan based on the each aircraft state to generate a corrected flight plan, and determining a portion of the corrected flight plan that remains to be flown; receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof; receiving weather data; generating predicted flight traffic data by performing a simulation of the flight over the duration of time, the simulation based at least in part on the consolidated data, the operational context data, and the weather data; and generating a user output based at least in part on the predicted flight traffic data.

Clause 2: the method of clause 1, further comprising: receiving user input representing the duration and a region of interest; and filtering the set of current flight plans based on the duration and the region of interest.

Clause 3: the method of clause 2, wherein the area of interest is associated with an airport terminal, with an airport, with multiple airports, or with a geographic area.

Clause 4: the method of clause 1, wherein each flight plan of the set of current flight plans includes an indication of a route to be followed, an aircraft type, an origin-destination pair, a predicted departure time, a predicted arrival time, or any combination thereof.

Clause 5: the method of clause 1, wherein the monitoring data comprises an auto dependent monitoring broadcast (ADS-B) report.

Clause 6: the method of clause 5, further comprising: storing at least one ADS-B report; and calculating flight trajectories based on the ADS-B reports, wherein at least one of the corrected flight plans is based on the flight trajectories.

Clause 7: the method of clause 1, further comprising: formatting the weather data to be compatible with the simulated architecture.

Clause 8: the method of clause 1, wherein the weather data is formatted into a weather information file, wherein the operational context data is formatted into a waypoint and navigation assistance file, an airport file, or both, and wherein the consolidated data is formatted into a flight schedule file.

Clause 9: the method of clause 1, wherein the simulation is performed using a fast time simulation tool.

Clause 10: the method of clause 1, wherein the user output comprises a predicted average delay for a region of interest at a time within the duration, a suggested action to reduce the predicted average delay, or both.

Clause 11: a system, the system comprising: at least one processor; and one or more memories storing instructions that, when executed by the processors, cause the at least one processor to: receiving flight plan data representative of a set of current flight plans; receiving monitoring data representative of a set of current aircraft states; generating consolidated data representative of a set of remaining flight portions to be flown, said consolidated data being generated by: pairing each aircraft state with each flight plan, correcting the each flight plan based on the each aircraft state to generate a corrected flight plan, and determining a portion of the corrected flight plan that remains to be flown; and sending the merged data to a simulator module.

Clause 12: the system of clause 11, wherein the instructions further cause the processor to: receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof; receiving weather data; and sending the operational context data and the weather data to the simulator module.

Clause 13: the system of clause 12, wherein the simulator module is configured to: generating predicted flight traffic data by performing a simulation of a flight of a duration of time, the simulation based at least in part on the consolidated data, the operational context data, and the weather data; and outputting at least a portion of the predicted flight traffic data to an output device.

Clause 14: the system of clause 11, wherein the instructions further cause the processor to: receiving user input representing a duration and a region of interest; and filtering the set of current flight plans based on the duration and the region of interest.

Clause 15: the system of clause 11, wherein the instructions further cause the processor to: storing the at least one ADS-B report in a memory; and calculating flight trajectories based on the ADS-B reports, wherein at least one of the corrected flight plans is based on the flight trajectories.

Clause 16: the system of clause 12, wherein the instructions further cause the processor to: formatting the weather data to be compatible with an architecture of the simulator module.

Clause 17: a method, the method comprising: receiving flight plan data representative of a set of current flight plans; receiving monitoring data representative of a set of current aircraft states; generating consolidated data representing a set of remaining portions of the flight to be flown based on the flight plan data and based on the monitoring data; receiving operational context data representative of an airspace configuration, an airport configuration, or a combination thereof; receiving weather data; iteratively changing parameters of the operational context data by performing a plurality of simulations of a set of flights in a duration, the plurality of simulations based at least in part on the consolidated data, the operational context data, and the weather data, and calculating, for each iteration, a predicted delay for the set of flights; and identifying, based on the plurality of simulations, a recommendation associated with the parameter to reduce the predicted delay.

Clause 18: the method of clause 17, further comprising: receiving user input representing the duration and a region of interest; and filtering the set of current flight plans based on the duration and based on the region of interest.

Clause 19: the method of clause 18, wherein the area of interest is associated with an airport terminal, with an airport, with multiple airports, or with a geographic area.

Clause 20: the method of clause 17, further comprising: generating a user output comprising a predicted average delay for a region of interest at a time within the duration.

While various embodiments have been shown and described, the present disclosure is not limited thereto and will be understood to include all such modifications and variations as would be apparent to one skilled in the art.

Claims (10)

1. A method, comprising the steps of:

receiving flight plan data (124) representative of a set of current flight plans (310);

receiving monitoring data (127) representative of a set of current aircraft states (320);

generating consolidated data (268) representative of a set of remaining flight portions (360) to be flown, said consolidated data (268) being generated by: pairing each aircraft state (322-324) with each flight plan (312-314), correcting the each flight plan (312-314) based on the each aircraft state (322-324) to generate a corrected flight plan (352-354), and determining a portion (362-364) of the corrected flight plan (352-354) that remains to be flown;

receiving operational context data (126) representative of an airspace configuration, an airport configuration, or a combination thereof;

receiving weather data (128);

generating predicted flight traffic data (270) by performing a simulation of the flight over the duration (262), the simulation based at least in part on the consolidated data (268), the operational context data (126), and the weather data (128); and

generating a user output (280) based at least in part on the predicted flight traffic data (270).

2. The method of claim 1, further comprising the steps of:

receiving a user input (260) representing the duration (262) and a region of interest (264); and

filtering the set of current flight plans (310) based on the duration (262) and based on the region of interest (264).

3. The method of claim 1, wherein each flight plan (312-314) in the set of current flight plans (310) includes: an indication of a route to be followed, a type of aircraft, an origin-destination pair, a predicted departure time, a predicted arrival time, or any combination thereof.

4. The method of claim 1, wherein the monitoring data (127) comprises auto dependent monitoring broadcast (ADS-B) reports.

5. The method of claim 1, further comprising:

formatting the weather data (128) to be compatible with the simulated architecture.

6. The method of claim 1, wherein the weather data (128) is formatted as a weather information file, wherein the operational context data (126) is formatted as a waypoint and navigation assistance file, an airport file, or both, and wherein the consolidated data (268) is formatted as a flight schedule file.

7. The method of claim 1, wherein the simulation is performed using a fast time simulation tool.

8. The method of claim 1, wherein the user output (280) comprises: a predicted mean delay (282) for a region of interest (264) at a time within the duration (262), a suggested action (284) to reduce the predicted mean delay (284), or both.

9. A system, the system comprising:

at least one processor (222); and

one or more memories (224) storing instructions that, when executed by the processors, cause the at least one processor to:

receiving flight plan data (124) representative of a set of current flight plans (310);

receiving monitoring data (127) representative of a set of current aircraft states (320);

generating consolidated data (268) representative of a set of remaining flight portions (360) to be flown, said consolidated data (268) being generated by: pairing each aircraft state (322-324) with each flight plan (312-314), correcting the each flight plan (312-314) based on the each aircraft state (322-324) to generate a corrected flight plan (352-354), and determining a portion (362-364) of the corrected flight plan remaining to be flown; and

sending the merged data (268) to a simulator module (230).

10. The system of claim 9, wherein the instructions further cause the processor (222) to:

receiving operational context data (126) representative of an airspace configuration, an airport configuration, or a combination thereof;

receiving weather data (128); and

sending the operational context data (126) and the weather data (128) to the simulator module (230).

Images